3 ms·
I don't think it would be necessary to know a priori that something is unpredictable; failing to find a pattern after some number of observations should allow a
by nitrogen 12y ago
I don't think it would be necessary to know a priori that something is unpredictable; failing to find a pattern after some number of observations should allow an algorithm to determine that the event is not predictable by that algorithm.
- baddox 12y agoThe trouble is that you can't reliably distinguish random data (or data determined by factors you're not considering) from patterned data.
- thaumasiotes 12y agoHow do you think random number generators are tested?